Superimposed Handwriting Recognition via Concurrent Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing on-line handwriting recognition systems face challenges in accurately interpreting natural handwriting with multi-stroke characters on small computing devices, where characters are written continuously without explicit separation, leading to overlaid or superimposed handwriting, which complicates determining character boundaries and reduces input efficiency.
Innovation Solution
The system performs concurrent fragmentation, segmentation, recognition, and interpretation using experts like a fragmentation expert, segmentation expert, recognition expert, and language expert through dynamic programming to analyze input strokes, classify fragments, and generate character, word, and sentence candidates, allowing for accurate recognition of superimposed and multi-stroke characters without explicit separation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multi-stroke recognition is used to recognize natural handwriting, then recognition accuracy for natural handwriting is improved, but determining character boundaries becomes more difficult
Solution Approach 1:
The patent applies segmentation by dividing the continuous handwriting input into discrete character units. The system segments superimposed strokes into individual character candidates by analyzing spatial overlap and temporal sequences, enabling multi-stroke characters to be properly separated and recognized without requiring explicit user separation gestures.
Solution Approach 2:
The patent performs preliminary fragmentation of input strokes into candidate characters before final recognition. By pre-processing the handwriting input to identify potential character boundaries and organize strokes into candidate groups, the system prepares the data structure needed for accurate multi-stroke character recognition while maintaining natural writing flow.
2Productivity
If characters are written continuously without explicit separation on small devices, then input efficiency is improved, but superimposed handwriting complicates character boundary determination
Solution Approach 1:
The patent employs dynamic algorithms that adapt to the varying patterns of superimposed handwriting. The system dynamically adjusts character boundary detection based on stroke overlap patterns, writing speed variations, and spatial relationships, allowing it to handle continuous input on small devices without requiring fixed or rigid boundary detection rules.
Solution Approach 2:
The patent introduces an intermediary processing layer that handles the complexity of superimposed handwriting. This intermediary system analyzes stroke sequences, identifies character candidate boundaries, and resolves ambiguities before presenting final character recognition results, thereby shielding the user from the underlying complexity while maintaining high input efficiency.
3Productivity
If concurrent fragmentation, segmentation, recognition, and interpretation are performed, then recognition speed and accuracy are improved, but system complexity increases
Solution Approach 1:
The patent merges multiple processing functions (fragmentation, segmentation, recognition, and interpretation) into a unified concurrent processing system. By integrating these functions to operate simultaneously rather than sequentially, the system achieves faster recognition speeds while managing complexity through shared data structures and coordinated processing pipelines that handle all functions in parallel.
Solution Approach 2:
The patent creates a multi-functional processing system where a single integrated engine performs fragmentation, segmentation, recognition, and interpretation tasks. This universal system uses common algorithms and data representations that serve multiple purposes, reducing overall system complexity compared to having separate dedicated systems for each function while maintaining high recognition speed and accuracy.
Data Source
AI summary
A system and method that is able to recognize a user's natural superimposed handwriting without any explicit separation between characters. The system and method is able to process single-stroke and multi-stroke characters. It can also process cursive handwriting. Further, the method and system can determine the boundaries of input words either by the use of a specific user input gesture or by detecting the word boundaries based on language characteristics and properties. The system and method analyzes the handwriting input through the processes of fragmentation, segmentation, character recognition, and language modeling. At least some of these processes occur concurrently through the use of dynamic programming.